Most companies that bought a CRM in recent years use it at 30%. The remaining 70% are features you pay for every month that were never switched on. Before you settle the first invoice, it's worth knowing what you actually want inside.
CRM is one of those tools where it's easiest to spend a lot of money with little effect. The demo looks spectacular, the salesperson promises "moving off Excel in two weeks", and three months later your team is still emailing each other because "doing it in the system is slower". This piece is about what to actually look at before you sign — and when you're better off not signing at all.
When a CRM actually makes sense
The mere fact that you have salespeople doesn't mean you need a CRM. A typical 12-person services company can work off a shared spreadsheet and an Outlook calendar for years — effectively. CRM only becomes necessary when one of the following starts being routine:
- A salesperson leaves and walks out with the customer history because "everything we knew about that client" lived in their inbox.
- Two people from the office call the same customer the same week because neither knew about the other.
- A customer says "you promised to come back to me on Friday" and your firm has no idea who was supposed to come back and with what.
- The owner can't answer "how many open quotes do we have right now" without a round of phone calls.
- Sales forecasts are produced on Friday evening on a "shout out your numbers, we'll merge them into one file" basis.
Each symptom alone is bearable. When three of five are present, ignoring the problem costs more than the CRM licence.
Three maturity levels
In Polish SMEs, CRMs broadly fall into three leagues — and picking the league matters more than picking the specific vendor.
Level 0 — shared spreadsheet. A well-named Excel or Google Sheets file with three tabs — contacts, quotes, tasks — is enough for a 5–15 person firm where everyone sees each other in the corridor. Cost: zero plus a disciplined owner who enforces the entries.
Level 1 — lightweight CRM. Pipedrive, Bitrix24, HubSpot Free, monday.com, Livespace, Firmao — tools at 25–120 EUR per user per month offering a sales pipeline, email integration and basic reporting. This is the growth zone for most 15–50 person firms. Realistic rollout in 4–6 weeks if someone actually sits down to do the work.
Level 2 — full enterprise CRM. Microsoft Dynamics 365, Salesforce, SugarCRM, Comarch CRM, SAP Customer Experience. It costs many times more, lets you configure deeply to your own processes, but requires an implementation team and an admin. In a 30-person manufacturing firm with Salesforce shoe-horned in, most of the team uses it like a worse Excel.
In 90% of cases, a company that runs on Excel today should stop at Level 1. Jumping to Level 2 only makes sense when you've genuinely used Level 1 for at least two years and you know exactly why it isn't enough anymore.
A real example — what a CRM actually delivered
At one of our clients — a 28-person distributor of crop protection products serving agricultural shops across southern Poland — sales by "the office team" (five people) was 100% run from Outlook and a single Excel file with 14 tabs. Enquiries came in by phone, email and occasionally WhatsApp from the field rep. A customer would say "I need a quote like the one from March" — and the office spent 40 minutes hunting for "what we had in March".
Instead of "rolling out Salesforce", we set them up on Pipedrive (about 12 EUR per user per month, around 75 EUR total), integrated with Comarch Optima via its public API and synced with Office 365 mailboxes. Configuration took six weeks — most of that went into agreeing what the sales funnel actually looks like (up to that point, nobody in the firm had written it down).
What changed after three months:
- Average response time to an enquiry fell from 36 hours to 4 hours — mostly because enquiries went into a shared queue rather than the private inbox of Mrs Jola, who was on holiday that week.
- The "I forgot to call back" count — previously totted up from complaints, around 11 per week — dropped to 2 per week.
- Quote-to-order conversion went from 19% to 26%. No magic. The team simply stopped losing leads they "hadn't got around to following up".
- The owner saw a "what's in progress" report for the first time in his life without picking up the phone.
Total project cost — annual licence plus rollout with the integration — under 6,500 EUR. That's the money one lost opportunity at a key account pays back in a single quarter.
What really has to work
In our experience, 80% of CRM value in an SME comes from very boring, basic functions. Before you compare "intelligent predictions" and "AI recommendations", make sure the tool nails the following:
- Email sync — Outlook or Gmail. Without it, the salesperson will be retyping emails into the CRM, which means they'll stop after two days.
- Shared customer card — with contact history, quotes and invoices. Ideally integrated with the ERP (Comarch Optima, Subiekt GT, enova365, Microsoft Dynamics).
- A sales pipeline with clear stages — so everyone sees what's "in proposal", "in negotiation", "closed".
- Tasks with reminders — built into the customer card, not in a separate app.
- Export to Excel — because there will always be a week when the built-in reports aren't enough.
- GDPR and EU hosting — either a Polish vendor or a vendor with an EU processing agreement. Without this, the lawyer says no.
If the salesperson is showing you "AI assistant" but can't answer how the sync with Comarch Optima works — go back to the must-have list. AI from your salesperson's point of view means nothing as long as the customer card doesn't pull in their last invoice numbers on its own.
Selection pitfalls
The recurring mistakes that cost firms the most:
- Buying features for later. "Let's take the Enterprise plan because maybe one day we'll have a marketing department." No. Buy the plan for what you need today and upgrade in a year.
- Picking a vendor without Polish ERP integration. Most Western CRMs put the Polish market in row three. Check who's actually going to integrate Optima for you — and what it costs.
- Rollout with no process owner. If "the CRM belongs to IT", it dies. The CRM must belong to the person responsible for sales.
- Migrating all the history at once. Move new customers and open cases; leave the history in an archive. Migrating eight years of data delays the rollout by six months.
- Skipping training. Two hours of instruction after go-live isn't enough. Plan 4–6 short, repeated sessions in the first quarter.
The first 90 days
A CRM rollout doesn't end on go-live day — it ends after three months of real use. During that window, watch three numbers: how many records get added weekly (should rise), how many pipeline items sit there with no recent note (should fall), and how many people on the team log in daily (should reach 100%).
If 30% of the team is still working from the old Excel on the side after 90 days — the rollout failed. Don't buy another CRM. Go back to the conversation about the sales process.
If you'd like to talk through whether your company is ready for a CRM, what maturity level fits, and what you actually need to integrate with Optima — the first call is free and takes half an hour.
A CRM won't sell for you. It just keeps you from forgetting the customers who want to buy from you.